Re: gEDA-user: Foss-pcb Proposed plan from CERN

[John Doty <jpd@xxxxxxxxx>]

On Aug 24, 2011, at 8:33 AM, Jared Casper wrote:

> I chuckled at what this community would think of the comment, in
> response to "There are users who prefer separate dedicated
> applications to an integrated design environment.", "BTW. How many of
> these users have ever designed a PCB with more than 4 layers and, say,
> 300 components? From my own experience, above the certain level of PCB
> complexity the intuitiveness and efficiency of the GUI become a
> paramount. "

I think that's exactly backwards. The "intuitiveness and efficiency" of the GUI make for comfort, but not productivity. In a big design, the key is to break it down into modules, and then use the automation to put the modules together. This is especially true when you recognize that a big design encompasses not just EDA, but documentation, software, and possibly other things. The toolkit approach allows you to combine these things in a maximally automated flow.

I've seen the difference starkly in software. I personally don't care what tools a programmer uses as long as they get the job done: this should be a matter of individual preference. Except, it is my experience that programmers who prefer toolkits are much more productive than programmers who prefer IDE. They plan better, they factor better, and they exploit the power of the computer better. One serious problem is that IDE encourages very inefficient debugging practices: it's much better to trap bugs with assertions, logs, and analysis than to fish for bugs interactively.

Yes, it takes more thought and planning to use a toolkit. For simple jobs, a nice intuitive GUI is fine (I'm typing this to the Mac "Mail" app). But planning is more important for big jobs, and a toolkit rewards planning better. Spending time to adapt your processes to the job is a big time saver for big jobs.

A flexible, extensible, toolkit is especially superior for jobs that have characteristics that fall outside the limits of the application designers' imaginations. Try exporting KiCad designs to a computer algebra system for symbolic analysis (but the Mathematica back end for gnetlist only took me an afternoon to write).

The important thing to recognize is that there is room for, and a need for, both toolkits and integrated tools. AWK and spreadsheets are both effective at processing tabular data in their own ways, but a merged tool with the characteristics of both would be incomprehensible. I think the same is true in EDA.

It is my opinion that gEDA's developers and users should embrace its strengths as a powerful, flexible toolkit. Keep the tools separate. Keep the interfaces clean and simple. Maximize the rewards that those who can do a little scripting can earn. Let KiCad cover the integrated app space.

It would be useful to be able to import KiCad schematics, so that when users are ready for the more powerful toolkit we could offer them an upgrade path.

John Doty              Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd@xxxxxxxxx




_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user